Generating capacity combination prediction method for new energy power generation with multi-algorithm access

Abstract

The invention relates to the technical field of new energy power generation capacity prediction, and discloses a multi-algorithm access power generation capacity combination prediction method for new energy power generation. The multi-algorithm access generating capacity combination prediction method for new energy power generation comprises the following steps: establishing a wind power plant comprehensive model; establishing a data storage comparison system; establishing an optimization combination model; according to the method, the wind power plant comprehensive model is combined with meteorological observation data and weather forecast data, so that the working environment of the virtual wind power plant is closer to reality, the prediction precision is further improved, the optimized combination model utilizes the advantages of two prediction technologies, the defects of a single prediction technology are improved, and the wind power plant comprehensive model is predicted and combined with external environment conditions and historical power generation conditions. On the basis of improving the prediction precision of the wind power plant comprehensive model and the prediction precision of the optimization combination model, the two prediction methods are combined to lead out the comprehensive optimization prediction generating capacity, so that the generating capacity prediction precision is further improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of renewable energy power generation prediction, and specifically relates to a combined power generation prediction method for renewable energy power generation with multiple algorithms accessed. Background Art

[0002] The increasingly serious global energy crisis makes the development of renewable energy more and more important. As a new energy, wind energy has the advantages of clean and renewable energy, and is increasingly valued by countries around the world. Wind power generation is characterized by randomness and volatility, and large-scale access to the power grid has brought impacts on the safe and stable operation of the power system. Predicting the future output power of wind power in advance and reserving the absorption space according to the prediction results are important technical means for my country to improve the absorption level of new energy and ensure the safety of the power system. The prediction of new energy power is affected by many factors. Any single model prediction model will have large prediction errors in certain scenarios due to its own limitations; therefore, it needs to be improved in view of the current situation. Summary of the invention

[0003] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides a combined prediction method for power generation of renewable energy with multi-algorithm access, which effectively solves the problem that the prediction of renewable energy power is affected by multiple factors, and any single model prediction model will have large prediction errors in certain scenarios due to its own limitations.

[0004] To achieve the above object, the present invention provides the following technical solution: a method for forecasting the combined power generation of renewable energy power generation using multiple algorithms, characterized in that it comprises the following steps:

[0005] Step 1: Establish a comprehensive model of the wind farm. Use modeling software to build a virtual model of the wind farm based on the terrain around the wind farm and the wind farm building drawings, combined with the wake effect of each unit in the wind farm, the shear effect of the surface landform and the tower shadow effect. At the same time, combine the physical engine to simulate the real weather changes and the changes of the terrain and vegetation around the wind farm in the four seasons, so as to build a comprehensive model of the wind farm. Use the comprehensive model of the wind farm in combination with meteorological observation data and weather forecast data to predict the power generation of the wind farm and obtain the predicted power generation of the comprehensive model of the wind farm;

[0006] Step 2: Establish a data storage and comparison system to collect and store the actual power generation data and environmental data of the wind farm, as well as the power generation data and environmental data predicted by the comprehensive model of the wind farm proposed in step 1, compare the actual data with the predicted data, and calculate the prediction deviation value;

[0007] Step three: Establish an optimized combination model. The optimized combination model is optimized and combined by weighted method using grey model and artificial neural network. Use MATLAB software to program and run to calculate the expected power generation of the optimized combination model. Then, compare it with the actual power generation through step two to obtain the predicted deviation value of the optimized combination model. Combine the predicted deviation value of the wind power plant comprehensive model and the predicted deviation value of the optimized combination model through the weight algorithm to calculate the weight value of the wind power plant comprehensive model and the weight value of the optimized combination model. Multiply the predicted power generation of the wind power plant comprehensive model by the weight value of the wind power plant comprehensive model plus the expected power generation of the optimized combination model by the weight value of the optimized combination model to obtain the comprehensive optimized predicted power generation.

[0008] Preferably, the modeling software proposed in step 1 is Unity3D, which can produce 3D games, VR models, three-dimensional animations and provide visualization operations. It has a built-in terrain editor and can freely establish terrain models. The physical engine proposed in step 1 is Unreal Engine, which has excellent visual effects and graphics rendering capabilities, supports advanced ray tracing technology, and has powerful physical simulation capabilities, supporting highly precise physical simulation.

[0009] Preferably, the grey model proposed in step three is a method of establishing a mathematical model and making predictions through a small amount of incomplete information. The grey model is based on the past and present development laws of objective things, and describes and analyzes future development trends and conditions through scientific methods to form scientific assumptions and judgments. The grey prediction model is suitable for situations where information is incomplete or the amount of data is small. It identifies the degree of difference in development trends between system factors, conducts correlation analysis, generates and processes the original data, finds the law of system changes, generates a data sequence with strong regularity, and then establishes a corresponding differential equation model to predict the future development trend of things.

[0010] Preferably, the artificial neural network proposed in step three is a new information processing system that imitates the structure and function of brain cells, the structure of brain nerves, and thinking processing problems. The artificial neural network analyzes a batch of mutually corresponding data provided in advance, grasps the potential rules between the two, and predicts new input data based on the potential rules.

[0011] Preferably, the weighting method proposed in step three refers to a method of assigning different proportions to various factors or variables according to their different importance or influence during the calculation or analysis process, and then performing weighted calculation.

[0012] Preferably, the wake effect proposed in step 1 refers to the phenomenon that when the wind turbines are in operation, the rotation of the rotor of each wind turbine modifies the flow wind field, so that the subsequent wind turbines are affected by the previous wind turbines and the wind speed is reduced and the wind resistance is increased.

[0013] Preferably, the surface landform shear effect proposed in step 1 refers to the phenomenon that, during the formation of the surface landform, due to crustal movement or external force, the surface material is deformed and displaced under the action of shear force.

[0014] Preferably, the tower shadow effect proposed in step one refers to the phenomenon of reduced wind speed due to the blocking effect of the tower, which is a negative effect that occurs during the power generation process of the wind turbine. When the wind flows through the tower, the tower will interfere with the airflow flowing through the blades, causing the direction and size of the wind vector to change. Especially in downwind wind turbines, the wind speed behind the tower will decrease, thereby affecting the output and power generation performance of the wind turbine.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. The present invention, by using a comprehensive model of a wind farm combined with meteorological observation data and weather forecast data, can simulate the atmospheric circulation and topography, the influence of buildings on wind, and the geographical location and height of the wind farm, so that the working environment of the virtual wind farm is closer to reality, thereby improving the prediction accuracy.

[0016] 2. The present invention optimizes the combined model by effectively combining the grey model and the artificial neural network in a weighted manner, utilizing the advantages of the two prediction technologies and improving the defects of a single prediction technology. It not only uses the prediction of external environmental conditions but also combines the historical power generation conditions to make the prediction more comprehensive, thereby improving the prediction accuracy.

[0017] 3. The present invention, on the basis of improving the prediction accuracy of the wind power plant comprehensive model and the prediction accuracy of the optimized combination model, combines the two prediction methods to derive a comprehensive optimized prediction of power generation, thereby further improving the prediction accuracy of power generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0019] In the attached picture:

[0020] Figure 1 This is a framework diagram of a method for forecasting the combined power generation of renewable energy using multiple algorithms of the present invention; DETAILED DESCRIPTION

[0021] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0022] See also Figure 1 The present invention provides a method for forecasting the combined power generation of renewable energy power generation using multiple algorithms, comprising the following steps:

[0023] Step 1: Establish a comprehensive model of the wind farm. Use modeling software to build a virtual model of the wind farm based on the terrain around the wind farm and the wind farm building drawings, combined with the wake effect of each unit in the wind farm, the shear effect of the surface landform and the tower shadow effect. At the same time, combine the physical engine to simulate the real weather changes and the changes of the terrain and vegetation around the wind farm in the four seasons, so as to build a comprehensive model of the wind farm. Use the comprehensive model of the wind farm in combination with meteorological observation data and weather forecast data to predict the power generation of the wind farm and obtain the predicted power generation of the comprehensive model of the wind farm;

[0024] Step 2: Establish a data storage and comparison system to collect and store the actual power generation data and environmental data of the wind farm, as well as the power generation data and environmental data predicted by the comprehensive model of the wind farm proposed in step 1, compare the actual data with the predicted data, and calculate the prediction deviation value;

[0025] Step three: Establish an optimized combination model. The optimized combination model is optimized and combined by weighted method using grey model and artificial neural network. Use MATLAB software to program and run to calculate the expected power generation of the optimized combination model. Then, compare it with the actual power generation through step two to obtain the predicted deviation value of the optimized combination model. Combine the predicted deviation value of the wind power plant comprehensive model and the predicted deviation value of the optimized combination model through the weight algorithm to calculate the weight value of the wind power plant comprehensive model and the weight value of the optimized combination model. Multiply the predicted power generation of the wind power plant comprehensive model by the weight value of the wind power plant comprehensive model plus the expected power generation of the optimized combination model by the weight value of the optimized combination model to obtain the comprehensive optimized predicted power generation.

[0026] Among them, the modeling software proposed in step 1 is Unity3D. Unity3D can make 3D games, VR models, 3D animations and provide visual operations. It has a built-in terrain editor and can freely build terrain models. The physical engine proposed in step 1 is Unreal Engine, which has excellent visual effects and graphics rendering capabilities, supports advanced ray tracing technology, and has powerful physical simulation capabilities and supports highly accurate physical simulation. The gray model proposed in step 3 is a method of building mathematical models and making predictions through a small amount of incomplete information. The gray model is based on the past and present development laws of objective things. It describes and analyzes future development trends and conditions through scientific methods to form scientific assumptions and judgments. The gray prediction model is suitable for situations where information is incomplete or the amount of data is small. By identifying the degree of difference in development trends between system factors, correlation analysis is performed, and the original data is generated and processed to find the laws of system changes and generate data sequences with strong regularity. Then, the corresponding differential equation model is established to predict the future development trend of things. The artificial neural network proposed in step 3 is a new type of neural network that imitates the structure and function of brain cells, the structure of brain nerves, and the way of thinking and processing problems. Information processing system, artificial neural network analyzes a batch of corresponding data provided in advance, grasps the potential rules between the two, and predicts new input data based on the potential rules. The weighted method proposed in step 3 refers to the method of assigning different proportions to various factors or variables according to their importance or influence during the calculation or analysis process, and then performing weighted calculation. The wake effect proposed in step 1 refers to the fact that when the wind turbine is in operation, the rotation of the rotor of each wind turbine will modify the wind field, causing the subsequent wind turbine to be affected by the previous wind turbine and reduce the wind speed. , the phenomenon of increasing wind resistance. The surface landform shear effect proposed in step one refers to the phenomenon that in the process of surface landform formation, due to crustal movement or external force, the surface material is deformed and displaced under the action of shear force. The tower shadow effect proposed in step one refers to the phenomenon of reduced wind speed caused by the blocking effect of the tower. It is a negative effect of wind turbines in the power generation process. When wind flows through the tower, the tower will interfere with the airflow flowing through the blades, causing the direction and size of the wind vector to change, especially in downwind wind turbines. The wind speed behind the tower will decrease, thereby affecting the output and power generation performance of the wind turbine.

[0027] By using the comprehensive model of the wind power plant in combination with meteorological observation data and weather forecast data, the working environment of the virtual wind farm can be made closer to reality by simulating the atmospheric circulation and terrain, the impact of buildings on wind, and the geographical location and height of the wind farm, thereby improving the prediction accuracy. The principle of the optimized combination prediction technology is simple and easy to understand, and it is also convenient to practice; the optimized combination model effectively combines the grey model and the artificial neural network in a weighted manner, utilizing the advantages of the two prediction technologies and improving the defects of a single prediction technology. It not only uses the prediction of external environmental conditions, but also combines historical power generation conditions to make the prediction more comprehensive, thereby improving the prediction accuracy; on the basis of improving the prediction accuracy of the comprehensive model of the wind power plant and the prediction accuracy of the optimized combination model, the two prediction methods are combined to derive a comprehensive optimized prediction of power generation, thereby further improving the prediction accuracy of power generation.

[0028] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0029] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A combined prediction method for power generation of renewable energy using multiple algorithms, characterized by: The following steps are involved: Step 1: Establish a comprehensive model of the wind farm. Use modeling software to build a virtual model of the wind farm based on the terrain around the wind farm and the wind farm building drawings, combined with the wake effect of each unit in the wind farm, the shear effect of the surface landform and the tower shadow effect. At the same time, combine the physical engine to simulate the real weather changes and the changes of the terrain and vegetation around the wind farm in the four seasons, so as to build a comprehensive model of the wind farm. Use the comprehensive model of the wind farm in combination with meteorological observation data and weather forecast data to predict the power generation of the wind farm and obtain the predicted power generation of the comprehensive model of the wind farm; Step 2: Establish a data storage and comparison system to collect and store the actual power generation data and environmental data of the wind farm, as well as the power generation data and environmental data predicted by the comprehensive model of the wind farm proposed in step 1, compare the actual data with the predicted data, and calculate the prediction deviation value; Step three: Establish an optimized combination model. The optimized combination model is optimized and combined by weighted method using grey model and artificial neural network. Use MATLAB software to program and run to calculate the expected power generation of the optimized combination model. Then, compare it with the actual power generation through step two to obtain the predicted deviation value of the optimized combination model. Combine the predicted deviation value of the wind power plant comprehensive model and the predicted deviation value of the optimized combination model through the weight algorithm to calculate the weight value of the wind power plant comprehensive model and the weight value of the optimized combination model. Multiply the predicted power generation of the wind power plant comprehensive model by the weight value of the wind power plant comprehensive model plus the expected power generation of the optimized combination model by the weight value of the optimized combination model to obtain the comprehensive optimized predicted power generation.

2. The method for forecasting the combined power generation of renewable energy with multiple algorithms accessed according to claim 1 is characterized by: The modeling software proposed in the step 1 is Unity3D, which can produce 3D games, VR models, three-dimensional animations and provide visual operations. It has a built-in terrain editor and can freely create terrain models. The physical engine proposed in the step 1 is Unreal Engine, which has excellent visual effects and graphics rendering capabilities, supports advanced ray tracing technology, and has powerful physical simulation capabilities, supporting highly precise physical simulation.

3. The method for forecasting power generation combined with renewable energy using multiple algorithms according to claim 1 is characterized by: The grey model proposed in the step three is a method of establishing a mathematical model and making predictions through a small amount of incomplete information. The grey model is based on the past and present development laws of objective things, and describes and analyzes future development trends and conditions through scientific methods to form scientific assumptions and judgments. The grey prediction model is suitable for situations where information is incomplete or the amount of data is small. It identifies the degree of difference in development trends between system factors, conducts correlation analysis, generates and processes the original data, finds the law of system changes, generates a data sequence with strong regularity, and then establishes a corresponding differential equation model to predict the future development trend of things.

4. The method for forecasting the combined power generation of renewable energy using multiple algorithms according to claim 1 is characterized by: The artificial neural network proposed in step three is a new type of information processing system that imitates the structure and function of brain cells, the structure of brain nerves, and thinking processing problems. The artificial neural network analyzes a batch of corresponding data provided in advance, grasps the potential rules between the two, and predicts new input data based on the potential rules.

5. The method for forecasting the combined power generation of renewable energy using multiple algorithms according to claim 1 is characterized by: The weighting method proposed in step three is a method of assigning different proportions to various factors or variables according to their importance or influence during the calculation or analysis process, and then performing weighted calculations.

6. The method for forecasting the combined power generation of renewable energy using multiple algorithms according to claim 1 is characterized by: The wake effect mentioned in step 1 refers to the phenomenon that when the wind turbines are in operation, the rotation of the rotor of each wind turbine will modify the wind field, so that the subsequent wind turbines are affected by the previous wind turbines and the wind speed is reduced and the wind resistance is increased.

7. The method for forecasting the combined power generation of renewable energy using multiple algorithms according to claim 1 is characterized by: The surface landform shear effect proposed in step 1 refers to the phenomenon that, during the formation of surface landforms, due to crustal movement or external forces, surface materials are deformed and displaced under the action of shear force.

8. The method for forecasting the combined power generation of renewable energy using multiple algorithms according to claim 1 is characterized by: The tower shadow effect mentioned in step 1 refers to the phenomenon of reduced wind speed due to the blocking effect of the tower, which is a negative effect of the wind turbine during the power generation process. When the wind flows through the tower, the tower will interfere with the airflow flowing through the blades, causing the direction and size of the wind vector to change. Especially in downwind wind turbines, the wind speed behind the tower will decrease, thereby affecting the output and power generation performance of the wind turbine.

Images